AI GPU Arch Perf Optimization Intern

Job Details:

Job Description:

The Role and Impact

As an AI Software Engineering Graduate Intern, you will play a key role in advancing the capabilities of Intel's GPU platforms by optimizing GPU compute kernels and validating GPU architectures using real AI workloads. Your contributions will directly impact hardware/software codesign and shape the next generation of Intel GPU and AI accelerator platforms while providing hands-on experience in GPU architecture and performance engineering.

Business Group

The Data Center Group (DCG) is integral to Intel's mission of advancing computing and connectivity at scale. This group focuses on creating innovative solutions for data center environments, including processors, accelerators, and infrastructure technologies to address the needs of AI, cloud computing, and high-performance computing. Joining DCG means contributing to the backbone of critical applications that power the digital world.

Key Responsibilities

- Analyze and optimize core GPU compute kernels for AI and numerical workloads (e.g., GEMM, Attention, operator fusion).

- Reproduce representative AI inference and training workloads for GPU IP validation.

- Perform GPU performance profiling and analysis to identify compute, memory, and pipeline bottlenecks.

- Build performance profiles and models to understand architecture-level performance behavior.

- Provide workload and kernel-level insights to support GPU architecture design and hardware/software codesign efforts.

Qualifications:

Minimum Qualifications

- Pursuing a Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, or a related technical field, with 0-1 years of hands-on experience in AI or GPU domains gained through internships, academic projects, coursework, or hands-on training.

- OR pursuing a Master's degree in Computer Science, Computer Engineering, Electrical Engineering, or a related technical field with no prior professional experience.

- Proficiency in Python for data analysis, experimentation, or tooling.

- Solid understanding of AI fundamentals, including common models and algorithms.

- Basic knowledge of computer systems (e.g., CPU/GPU architecture, memory systems, and performance analysis).

Preferred Qualifications

- Experience with GPU kernels or programming models such as CUDA, OpenCL, SYCL, or Triton.

- Exposure to performance optimization, compiler technologies, or parallel computing coursework, research, or internships.

- Strong analytical and problem-solving skills, with the ability to derive insights from profiling data.

- Interest in AI systems and infrastructure, beyond model-level development.

- Ability to work effectively in collaborative, cross-functional engineering teams.

We look forward to welcoming individuals who are passionate about shaping the future of AI and GPU architecture.

Job Type:

Student / Intern

Shift:

Shift 1 (China)

Primary Location:

PRC, Shanghai

Additional Locations:

PRC, Beijing

Posting Statement:

All qualified applicants will receive consideration for employment without regard to race, color, religion, religious creed, sex, national origin, ancestry, age, physical or mental disability, medical condition, genetic information, military and veteran status, marital status, pregnancy, gender, gender expression, gender identity, sexual orientation, or any other characteristic protected by local law, regulation, or ordinance.

Position of Trust

N/A

Work Model for this Role

This role will require an on-site presence. * Job posting details (such as work model, location or time type) are subject to change.

*

ADDITIONAL INFORMATION: Intel is committed to Responsible Business Alliance (RBA) compliance and ethical hiring practices. We do not charge any fees during our hiring process. Candidates should never be required to pay recruitment fees, medical examination fees, or any other charges as a condition of employment. If you are asked to pay any fees during our hiring process, please report this immediately to your recruiter.